A semantic description for content-based image retrieval

Robust and flexible semantic labeling of images is still a basic problem in content-based image representation and retrieval. In this paper, a self-organizing image description model (SID) was put forward for describing the image high-level semantic content. This model is a hierarchical architecture...

Full description

Saved in:

Bibliographic Details
Published in:	2008 International Conference on Machine Learning and Cybernetics Vol. 5; pp. 2466 - 2469
Main Authors:	Bing Wang, Xin Zhang, Ziao-Yan Zhao, Zhi-De Zhang, Hong-Xia Zhang
Format:	Conference Proceeding
Language:	English
Published:	IEEE 01-07-2008
Subjects:	Algorithm design and analysis Clustering algorithms Content-based image retrieval Feature extraction Image clusters Image retrieval Image semantic model Semantic description Support vector machines SVM Visualization
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Robust and flexible semantic labeling of images is still a basic problem in content-based image representation and retrieval. In this paper, a self-organizing image description model (SID) was put forward for describing the image high-level semantic content. This model is a hierarchical architecture, which includes primitive image layer, image feature layer, image semantic layer, multi-level semantic pattern layer and semantic labeling layer. A semantic-based retrieval algorithm (SBRA) for image high-level semantic content retrieval was designed and implemented. The performance of an experimental image retrieval system is evaluated on a database of around 3000 images. The experimental results show that SID and SBRA are effective in describing image high-level semantic content and can provide flexible image description and efficient image retrieval performance.
ISBN:	1424420954 9781424420957
ISSN:	2160-133X
DOI:	10.1109/ICMLC.2008.4620822